I
by
EDSON NCUBE
Dissertation presented in partial fulfilment of the requirements for the degree of
Doctor of Philosophy in the Faculty of AgriSciences at the University of Stellenbosch
Supervisor: Prof A. Viljoen
Co-supervisors: Prof B.C. Flett
Prof J. Van den Berg
March 201
7
The financial assistance of the National Research Foundation (NRF) towards this
research is hereby acknowledged. Opinions expressed and conclusions arrived at,
II
DECLARATION
By submitting this dissertation electronically, I declare that the entirety of the work contained therein is my own, original work, that I am the sole author thereof (save to the extent explicitly otherwise stated), that reproduction and publication thereof by Stellenbosch University will not infringe any third party and that I have not previously in its entirety or in part submitted it for obtaining any qualification.
Date: March 2017 Signature:
Copyright © 2017 Stellenbosch University All rights reserved
III
SUMMARY
Maize is a crop of great economic importance in southern Africa, and is widely consumed as a staple food and animal feed. Production of maize, however, is hampered by pathogens and pests such as Fusarium verticillioides and the African stem borer Busseola fusca, respectively.
Fusarium verticillioides infection results in Fusarium ear rot (FER) and contamination of maize
kernels with fumonisin mycotoxins, while B. fusca, causes significant damage to maize tissues during larval feeding. Despite attempts to control F. verticillioides, fungal infection and fumonisin production remains a threat to maize production due to a lack of resistant maize cultivars and the inability to target the pathogen with fungicides and biocontrol products. Planting Bt maize hybrids have become an important mechanism for the management of stem borers of maize. However, the recent discovery of B. fusca resistance to Bt maize with a single crystal protein MON810 gene, indicates that care should be taken not to solely rely on this technology for the management of B. fusca.
The interactive effect of B. fusca and F. verticillioides on FER and fumonisin production in maize was investigated in this study. Maize ears were inoculated with F. verticillioides alone, with both F. verticillioides and B. fusca, and with B. fusca alone. Fusarium verticillioides isolate MRC826 was inoculated by injecting a spore suspension of the fungus into the silk channel of each primary ear at the blister stage. For B. fusca infestation, aliquots of 10-15 neonate larvae were deposited into the whorl of each plant at the 12-13th leaf stagebefore tasselling using a mechanical applicator. Maize ears were also mechanically wounded at the blister stage with a cork borer (different sizes and number of wounds) to mimic hail damage, and half of the wounds infected with F. verticillioides. Results from this study indicated that the impact of B.
fusca infestation on FER varied seasonally, possibly due to its sporadic damage to maize
ears. Busseola fusca, however, did not result in a significant increase in fumonisin production. The severity of wounding of maize ears was an important contributor to FER development and fumonisin production.
The effect of host plant genetic modification and pesticide application on FER and fumonisin production in maize was investigated by studying the response of a Bt hybrid and its non-Bt isohybrid to F. verticillioides infection and B. fusca infestation; and by treating plants with Beta-cyfluthrin (non-systemic) and Benfuracarb (systemic) insecticides. The field trials were conducted over three seasons using a randomised complete block design with six replicates per treatment. Uninoculated, uninfested and undamaged control treatments were included. All ears were harvested at physiological maturity and FER, total fumonisin concentration, stem borer cumulative tunnel length(B. fusca damage) and target DNA of fumonisin-producing Fusarium spp. were quantified. Busseola fusca infestation had no effect
IV
on fungal colonisation and fumonisin production in maize. Bt and non-Bt kernels were equally contaminated with fungal DNA, but FER and fumonisin production were reduced in the Bt hybrid under natural farming conditions. Despite the evidence found in this study and others that Bt maize indirectly reduces FER and fumonisin production, this was also inconsistent over seasons. Benfuracarb controlled stem borers, and therewith indirectly reduced FER and fumonisin production. FER development and fumonisin production by F. verticillioides varied over seasons, indicating the importance of environmental conditions on FER and fumonisin production.
A survey was also conducted at two sites in the North West province and one site in the Free State province of South Africa to analyse mycoflora in B. fusca frass. The exposure of B.
fusca larvae to F. verticillioides in stem borer frass was also evaluated in both greenhouse
and field trials. Maize whorls were inoculated with a spore suspension of F. verticillioides MRC826 4 weeks after plant emergence and infested with aliquots of 5-10 neonate B. fusca larvae 2 days later. The control treatment consisted of B. fusca infestation only. Several fungal species were associated with stem borer frass, including Acremonium zeae, Aspergillus
flavus, A. niger, F. chlamydosporum, F. incarnatum-equiseti species complex, F. oxysporum,
F. subglutinans, F. verticillioides, Mucor circinelloides, Rhizopus oryzae and Talaromyces flavus. The occurrence of A. niger in the frass suggests that further studies need to be
conducted to determine the effect of A. niger infection on fumonisin production in maize in South Africa. DNA quantity of fumonisin-producing Fusarium spp. was significantly more in frass collected from greenhouse plants inoculated with F. verticillioides and infested with B.
fusca larvae than in frass collected from the uninoculated and infested control, whilst the field
trial showed no significant differences in quantity of target DNA in frass from inoculated and uninoculated plants infested with B. fusca larvae. This indicates that plants in the field were naturally infected with F. verticillioides.
This study showed that Bt maize had no effect on infection of maize ears by fumonisin-producing Fusarium spp. and the subsequent production of fumonisin in F. verticillioides-inoculated maize ears, indicating that the effect of Bt maize on fumonisin production in maize ears is indirectly associated with its control of severe stem borer damage. Busseola fusca frass was a reservoir of different fungal species; some pathogenic to maize, and others antagonistic to maize pathogens. Moreover, B. fusca infestation of maize stems was associated with higher levels of fumonisin-producing Fusarium spp. in larval frass when F.
verticillioides was present on the plant. Multiple large wounds created by cork borers resulted
in significantly more FER symptoms and fumonisin production, irrespective of artificial F.
verticillioides inoculation of maize ears whereas B. fusca infestation resulted in a significant
increase in FER in only one of the three seasons, moreover, it had no effect on fumonisin production in all three seasons. This indicates that severe wounds that opens up husk
V
coverage and exposes maize kernels; caused by factors such as insects, hail and bird damage, and damage by implements; are important entry points for F. verticillioides that may lead to the transition from symptomless infection to necrotrophic pathogenicity resulting in FER and concomitant fumonisin production in maize kernels. However, climatic conditions are also important in FER and fumonisin production in maize. Moreover, Acremonium zeae endophytes occurring in frass can be used for the biological control of F. verticillioides resulting in the management of FER and subsequent fumonisin production.
VI
OPSOMMING
Mielies is ‘n gewas van groot ekonomiese belang in suidelike Afrika wat wyd as stapel voedsel en veevoer dien. Mielieproduksie word egter belemmer deur onderskeidelik patogene en peste soos Fusarium verticillioides en die Afrika stamboorder, Busseola fusca. Fusarium
verticillioides infeksie lei tot Fusarium kopvrot (FKV) en die produksie van fumonisien
mikotoksiene, terwyl die Afrika stamboorder, Busseola fusca, beduidende skade verrig aan mielieweefsel tydens larf voeding. Ondanks pogings om F. verticillioides te beheer, bly infeksie en fumonisienproduksie ‘n bedreiging vir mielieproduksie weens die tekort aan weerstandbiedende mieliekultivars en die onvermoë om die patogeen te teiken met swamdoders en biologiese beheerprodukte. Die aanplanting van Bt mieliebasters is ’n belangrike meganisme vir die bestuur van mieliestamboorders. Die onlangse ontdekking van
B. fusca weerstand teen Bt mielies met ‘n enkele kristal proteïen geen, dui egter daarop dat
daar nie slegs op hierdie tegnologie gesteun moet word vir die bestuur van B. fusca nie. Die interaktiewe effek van B. fusca en F. verticillioides op FKV en fumonisienproduksie in mielies is tydens hierdie studie ondersoek. Mieliekoppe is slegs F. verticillioides geïnokuleer, met beide F. verticillioides en B. fusca, en met slegs B. fusca. Fusarium verticillioides isolaat MRC826 is geïnokuleer deur ‘n spoorsuspensie van die fungus in te spuit in die sy kanaal van elke primêre kop tydens die blaas fase. Vir B. fusca infestasie, is afmetings van 10-15 neonaat larwe gedeponeer in die krans van elke plant by die 12-13de blaarfase voor pluimvorming, deur gebruik te maak van ‘n meganiese toediener. Mieliekoppe is ook meganies gewond tydens die blaasfase met ‘n kurkboorder (verskillende groottes en hoeveelhede wonde), om sodoende haelskade na te boots, en die helfde van die wonde is geïnfekteer met F.
verticillioides. Resultate van hierdie studie het aangedui dat die impak van B. fusca infestasie
op FKV seisoenaal variasie getoon het, moontlik weens die sporadiese skade aan mieliekoppe. Busseola fusca infestasie het egter nie gelei tot ‘n beduidende toename in fumonisienproduksie nie. Die graad van verwonding van die mieliekoppe was ‘n belangrike bydraer tot FKV ontwikkeling en fumonisienproduksie.
Die effek van gasheer plant genetiese verandering en plaagdoder toediening op FKV en fumonisienproduksie in mielies is ondersoek deur die reaksie te bestudeer van ‘n Bt kruising en sy nie-Bt iso-kruising tot F. verticillioides infeksie en B. fusca infestasie; en deur die behandeling van plante met Beta-cyfluthrin (nie-sistemies) en Benfuracarb (sistemies) insekdoders. Die veldproewe is oor ‘n tydperk van drie seisoene uitgevoer met die gebruik van ‘n ewekansige volledige blokontwerp met ses herhalings per behandeling. Ongeïnokuleerde, nie-geïnfesteerde en onbeskadigde kontrole behandelings is ingesluit. Alle koppe is geoes by fisiologiese volwassenheid en FKV, totale fumonisien konsentrasie,
VII
stamboorder kumulative tonnel lengte (B. fusca skade) en die hoeveelheid teiken DNA van fumonisien-produserende Fusarium spp. is gekwantifiseer. Busseola fusca infestasie het geen effek gehad op swam kolonisasaie en fumonisienproduksie in mielies nie. Bt en nie-Bt pitte is eweveel gekontamineer met swam DNA, maar FKV en fumonisienproduksie het afgeneem in die Bt kruising. Ten spyte van die bewyse gevind in hierdie en ander studies dat Bt mielies indirek FKV en fumonisienproduksie verminder, was dit ook strydig oor seisoene. Benfurakarb het stamboorders beheer, en saam met dit ook indirek FKV en fumonisienproduksie verminder. FKV ontwikkeling en fumonisienproduksie deur F. verticillioides het oor seisoene gevarieer, wat dui op die belangrikheid van omgewingstoestande vir FKV en fumonisienproduksie.
‘n Opname is ook by twee areas in die Noordwes Provinsie en een area in die Vrystaat Provinsie van Suid-Afrika uitgevoer, om mikoflora in B. fusca wurmboorsel te analiseer. Blootstelling van B. fusca larwe teenoor F. verticillioides in staboorder wurmboorsel is ook geëvalueer in beide kweekhuis- en veldproewe. Mieliekranse is met ‘n spoorsuspensie van F.
verticillioides MRC 826 geïnokuleer 4 weke na opkoms en 2 dae later geïnfesteer met
afmetings van 5-10 neonaat B. fusca larwe. Die kontrole behandeling het bestaan uit slegs B.
fusca infestasie. Verskeie swamspesies is geassosieer met stamboorder wurmboorsel,
insluitend Acremonium zeae, Aspergillus flavus, A. niger, F. chlamydosporum, F.
incarnatum-equiseti spesiekompleks, F. oxysporum, F. subglutinans, F. verticillioides, Mucor
circinelloides, Rhizopus oryzae en Talaromyces flavus. Die voorkoms van A. niger in die
wurmboorsel dui daarop dat verdere studies uitgevoer moet word om die effek van A. niger infeksie op fumonisienproduksie in mielies in Suid-Afrika te bepaal.
DNA hoeveelheid van fumonisien-produserende Fusarium spp. was betekenisvol hoër in wurmboorsel wat versamel is van kweekhuisplante geïnokuleer met F. verticillioides en geïnfesteer met B. fusca larwe as in wurmboorsel versamel vanaf die ongeïnokuleerde kontrole terwyl die veldproef geen betekenisvolle verskille in kwantiteit teiken DNA in wurmboorsel vanaf geïnokuleerde en nie-geïnokuleerde plante geïnfesteer met B. fusca larwe getoon het nie. Dit dui daarop dat plante in die veld natuurlik deur F. verticillioides geïnfekteer is.
Die studie het aangedui dat Bt mielies geen effek op infeksie van mieliekoppe het by fumonisien produserende Fusarium spp. en vervolgens fumonisienproduksie in geïnokuleerde
F. verticillioides mieliekoppe nie. Dit toon dat die effek van Bt mielies op fumonisienproduksie
in mieliekoppe indirek met die beheer van stamboorder skade geassosieer is. Hierdie studie het getoon dat B. fusca wurmboorsel as reservoir gedien het vir verskeie swamspesies; sommiges patogenies vir mielies, en ander antagonisties tot mieliepatogene. Met infestasie van B. fusca op mielie stamme word ondervind dat hoër vlakke van fumonisien produserende
VIII
verticillioides op plante. Verskeie groot wonde veroorsaak deur kurkboorders het gelei tot
beduidend meer FKV simptome en fumonisienproduksie, ongeag kunsmatige F. verticillioides inokulasie van mieliekoppe, terwyl B. fusca infestasie gelei het tot ‘n beduidende toename in FKV in slegs een van die drie seisoene. Verder het dit geen effek gehad op fumonisienproduksie in enige van die drie seisoene nie. Dit dui daarop dat wonde wat dopbedekking oopmaak en pitte blootstel; veroorsaak deur faktore soos insekte, haelskade, voëlskade, en implimentskade; belangrike toegangspunte is vir F. verticillioides wat mag lei tot die oorgang van simptoomlose endofitisme na nekrotrofiese patogenisiteit wat lei tot FKV en gepaardgaande fumonisienproduksie in mieliepitte. Klimaatstoestande is egter ook belangrik in FKV en fumonisienproduksie in mielies. Verder, kan Acremonium zeae endofiete wat in wurmboorsel voorkom vir die biologiese beheer van F. verticillioides gebruik word wat lei tot die bestuur van FKV en gevolglike fumonisienproduksie.
IX
ACKNOWLEDGEMENTS
I wish to express my sincere gratitude and appreciation to the following persons and institutions:
Prof A. Viljoen (Stellenbosch University), Prof B.C. Flett (ARC-GCI, North-West University) and Prof J. Van den Berg (North-West University), my supervisors who provided invaluable support and guidance;
Prof G.P. Munkvold (Iowa State University, USA), for assistance with drafting the project proposal;
Prof J.B.J. van Rensburg GCI), Dr M. Truter PPRI), Dr A. Erasmus (ARC-GCI), Dr B. Janse Van Rensburg (ARC-(ARC-GCI), Dr A. Schoeman (ARC-(ARC-GCI), Ms D. Biya (ARC-GCI), Mr O. Rhode (ARC-GCI) and Dr J du Toit (ARC-GCI) for assistance with
Busseola fusca rearing and supply, phylogenetic analysis, editing and statistical analyses;
Current and former personnel of the ARC-Grain Crops Institute, for technical assistance:
Laboratory: Ms M. Kwele, Dr B. Janse van Rensburg, Ms D. Biya, Ms M. Mahlobo, Ms S. Phokane, Mr A. Tantasi, Mr J. Viviers
Field work: Mr F. Mashinini, Mr J. Bass, Ms Y. Maila, Ms M. Du Toit, Ms S. Nthangeni, Ms K. Makokoe
Research articles: Ms J. Ramaswe, Ms J. Kilian;
The Agricultural Research Council, The Maize Trust and the National Research Foundation of South Africa, for financial support;
My wife Shushu, and my daughter Lachelle, and many others whom I have been in contact with during my PhD studies, including my friends at Prof Dr Hans-Ulrich Humpf’s laboratory, Institut für Lebensmittelchemie, Universität Münster, Germany.
X
CONTENTS
FULFILLMENT……….. I DECLARATION………. II SUMMARY/OPSOMMING………... III ACKNOWLEDGEMENTS……….... IXChapter 1: The Fusarium verticillioides x B. fusca interactions in maize: A
review……… 1
INTRODUCTION………... 1
THE FUSARIUM EAR ROT PATHOGEN: FUSARIUM VERTICILLIOIDES ……….. 2
Life cycle of F. verticillioides………... 3
Pathogenicity of F. verticillioides………... 4
Toxigenicity of F. verticillioides………... 4
Relationship between pathogenicity and toxigenicity……... 6
Factors modulating Fusarium ear rot and fumonisin production in maize……….... 6
Environmental conditions………... 6
Physio-chemical composition of kernels…..…... 7
Insects………. 7
Production practices………... 8
Management of Fusarium ear rot and fumonisins………….. 8
Biological methods…..……….. 8
Chemical methods………... 9
Cultural practices…..………. 9
Plant resistance ………. 10
Post-harvest storage and treatment..……….. 10
THE ARICAN MAIZE BORER BUSSEOLA FUSCA………... 11
Life cycle of B. fusca……….………... 12
Pupal and moth stage……….... 12
Egg and larval stage……….………. 12
Busseola fusca host plants………... 13
Behaviour of B. fusca moths and larvae on host plants……. 13
Damage caused to maize plants by stem borers……….. 15
XI
Environmental conditions……….. 16
Production practices………... 16
Management of maize stem borers………..…… 16
Biological control……….... 16
Chemical control ……….………..….... 17
Cultural practices…..………... 18
Push-pull habitat manipulation ……... 18
Plant resistance ……….……… 19
THE FUSARIUM VERTICILLIOIDES x BUSSEOLA FUSCA x MAIZE INTERACTION……… 20
Relationship of maize with F. verticillioides and B. fusca….. 20
Resistance in maize to F. verticillioides and B. fusca …..….. 21
Resistance mechanisms………... 21
Plant improvement .………... 24
CONCLUSIONS……….... 25
Chapter 2: The effect of Busseola fusca and mechanical wounding on Fusarium ear rot development and fumonisin production in maize……….. 57
ABSTRACT……… 57
INTRODUCTION………... 58
MATERIALS AND METHODS……… 60
RESULTS………... 64
DISCUSSION………. 66
Chapter 3: Fusarium ear rot and fumonisins in maize kernels when comparing a Bt hybrid with its non-Bt isohybrid and use of conventional insecticides to control Busseola fusca infestations……….……… 84
ABSTRACT……… 84
INTRODUCTION………... 85
MATERIALS AND METHODS……… 87
RESULTS………... 91
XII
Chapter 4: Mycoflora of Busseola fusca frass, and the possible role
of larvae in disseminating Fusarium verticillioides ……… 108
ABSTRACT……… 108
INTRODUCTION………... 109
MATERIALS AND METHODS……… 110
RESULTS………... 113
DISCUSSION………. 114
1
CHAPTER 1
The Fusarium verticillioides x Busseola fusca interactions in maize: A review
INTRODUCTION
Maize (Zea mays L.) originated in Mexico about 7 000 years ago (Ranum et al., 2014). Today the crop is produced throughout the world, with the United States, China and Brazil among the top three maize-producing countries, with approximately 65% of the total worldwide production in 2014/15 (USDA, 2016). South Africa is in the top ten maize-producing countries and was top-ranked in Africa in 2014/15 (USDA, 2016). In developed countries maize is processed into food, feed and industrial products that include starch, sweeteners, oil, beverages, glue, industrial alcohol and fuel ethanol. Approximately 40% of maize produced in the United States is now used for ethanol production (Ranum et al., 2014).
Maize is the staple food for many people in southern Africa and is a major constituent of animal feeds (SAGIS, 2014). In South Africa the crop is commercially cultivated in eight of the nine provinces, including the Free State, Mpumalanga, North West, Gauteng, KwaZulu-Natal, Limpopo and the Eastern and Northern Cape provinces. The bulk of the commercial maize production occurs within the Maize Triangle (De Waele and Jordaan, 1988), comprising of the western Free State, North West, Gauteng and western Mpumalanga provinces. Annual maize production by commercial farmers for the 2013/14 growing season was 14.3 million tonnes from 2.7 million hectares (ha) of land. Non-commercial maize production amounted to 675 000 tonnes from 408 000 ha (DAFF, 2014; SAGIS, 2014). Dry land farmers produced an average of 3.58 tons/ha while irrigation farmers produced an average of 10 tons/ha in the 2013/14 season (SAGIS, 2014).
One of the most damaging pathogens of maize is Fusarium verticillioides Sacc. Nirenberg (syn = F. moniliforme Sheldon), a fungus that is frequently associated with the crop in most production areas of the world (Desjardins, 2006; Ncube et al., 2011). Fusarium verticillioides is best known for causing Fusarium ear rot (FER), but the symptoms it cause can vary from non-symptomatic infections to severe rotting of roots, stems and ears (White, 1999). The most detrimental effect of F. verticillioides, however, is that it produces fumonisin mycotoxins that have been associated with diseases of humans and livestock (Marasas, 2001). Fusarium
verticillioides is also a pathogen of rice (Oryza sativa L.), sorghum (Sorghum bicolor L.) and
sugarcane (Saccharum officinarum L.) (McFarlane and Rutherford, 2005; Leslie and Summerell, 2006; McFarlane et al., 2009).
2
Maize production is also negatively affected by stem borer infestations (Porter et al., 1991; Kfir, et al., 2002; Hutchison et al., 2010). The African stem borer, Busseola fusca Fuller (Lepidoptera: Noctuidae), is indigenous to sub-Saharan Africa and causes damage to all plant parts in cultivated crops (Calatayud et al., 2014). In South Africa the insect is prevalent in the Highveld and the western maize production areas of the country (Kfir and Bell, 1993; Kfir, 2000; 2002), and can cause an estimated annual loss of 10-60% of total maize production (Kfir et al., 2002). Busseola fusca occurs under cool and wet conditions at altitudes ranging from sea level to 2 000 m above sea level (Abate et al., 2000), and is the most injurious pest of maize in South Africa (Annecke and Moran, 1982; Van Rensburg et al., 1988a). Busseola
fusca also causes damage to sorghum (Kfir et al., 2002), pearl millet (Pennisetum glaucum
L.) (Harris and Nwanze, 1992) and sugarcane (Assefa et al., 2015).
Planting of genetically modified maize hybrids (Bt maize) with Bacillus thuringiensis Berliner 1915 genes encoding for the δ-endotoxin crystal proteins that are toxic to lepidopteran insects has become an important method for the management of stem borers (Hellmich et al., 2008). However, most subsistence farmers in South Africa and all farmers in countries where restrictions on planting Bt maize are in place (Meissle et al., 2010), still rely on traditional pest management methods such as insecticides and residue management to control maize lepidopteran pests (Ncube, 2008). The interaction between B. fusca and F. verticillioides, however, is not sufficiently understood. This review, therefore, summarises available literature on the effect of stem borers such as B. fusca and F. verticillioides on FER and fumonisin production in maize.
THE FUSARIUM EAR ROT PATHOGEN: FUSARIUM VERTICILLIOIDES
Fusarium verticillioides was previously known as F. moniliforme, but its name has been
changed by Seifert et al. (2003) based on the fact that F. moniliforme represented an unacceptably broad species concept. Fusarium verticillioides was undisputedly the older name for the maize pathogen. Fusarium verticillioides has a sexual stage, which was previously known as Gibberella moniliformis Wineland (Leslie and Summerell, 2006).
Fusarium verticillioides is easily recognised by its cultural and morphological characteristics.
It produces white mycelia that turn violet when grown on potato dextrose agar. On carnation leaf agar it forms macroconidia that are relatively long and slender, as well as microconidia that are produced in long chains on monophialides (Leslie and Summerell, 2006). During the sexual stage, ascospores are produced in perithecia (Leslie and Summerell, 2006).
Ear rot diseases of maize can be produced by other Fusarium spp. too. These species include F. proliferatum Matsushima, Nirenberg; F. boothii O'Donnell, Aoki, Kistler & Geiser; F.
3
& Hansen; F. semitectum Berk. & Ravenel; F. subglutinans Wollenw. & Reinking, Nelson, Toussoun & Marasas (Summerell et al., 2003) and F. temperatum Scauflaire & Munaut (Schoeman, 2014; Zhang et al., 2014). Of these F. proliferatum, F. temperatum and F.
subglutinans cause FER, whereas F. graminearum and F. boothii are responsible for
Gibberella ear rot. Gibberella ear rot reduces maize yields and quality, and results in the contamination of maize ears with zearalenone and trichothecene mycotoxins (Desjardins, 2006). Other ear rots that occur in maize include Aspergillus ear rot and Diplodia ear rot. Aspergillus ear rot is caused by Aspergillus flavus Link ex Fr and Aspergillus parasiticus Speare (Gourama and Bullerman, 1995), which produce aflatoxin mycotoxins in maize (Diener
et al., 1987). Diplodia ear rot is caused by Stenocarpella maydis (Berk.), which results in
diplodiosis, a nervous disorder in cattle (Bos taurus L.) and sheep (Ovis aries L.) (Snyman et
al., 2011).
Life cycle of F. verticillioides
Kernel infection of maize is mainly due to airborne spores of F. verticillioides that infect through the silk channel (Munkvold and Carlton, 1997; Galperin et al., 2003) during the silking growth stage of the maize plant (Fig. 1). Airborne spores are produced as microconidia on the previous crop residue, and are disseminated by wind (Munkvold and Desjardins, 1997). After landing on the silks, the fungus progresses through the silk channel and into maize kernels. FER symptoms develop on ear tips or randomly as scattered areas of infection over the ear surface (Parsons and Munkvold, 2012). Direct invasion of kernels may also occur through stress cracks in the pericarp and through the pedicel (Odvody et al., 1997; Parsons and Munkvold, 2012), or by means of wounds caused by insects (Munkvold et al., 1997) and hail (Robertson et al., 2011). At the end of the maize-growing season, the fungus survives in maize residues as thickened hyphae on moist soils (Fig. 2) (Cotten and Munkvold, 1998; Munkvold, 2003). It does not produce chlamydospores, and thus cannot survive for extended periods (Leslie and Summerell, 2006).
Fusarium verticillioides that survives on maize residues at the end of a planting season is
sometimes released back into the soil (Munkvold and Desjardins, 1997). At the onset of a new maize-growing season, the soilborne hyphae germinate and grows into lateral roots (Oren et
al., 2003), after which the fungus colonises the stems of seedlings. Seeds, when contaminated
with F. verticillioides in the field, can also result in systemic infections of maize seedlings. This growth of the fungus may continue asymptomatically until seed production, when the fungus grows into the ear and colonise maize kernels (Fig. 2).
4 Pathogenicity of F. verticillioides
Fusarium verticillioides causes damping-off, stem malformation, as well as ear, stem and root
rot of maize (Nelson et al., 1993; Leslie and Summerell, 2006). Of these, FER is considered the most important. FER becomes visible as pink or white fungal growth on damaged and undamaged kernels (White, 1999). Some kernels do not become discoloured, but rather show white streaks with a starburst appearance (Duncan and Howard, 2010). Fusarium
verticillioides can also infect maize kernels without showing any visible symptoms (Oren et al.,
2003). Plant malformation manifests as twisted foliage and tillers (Cardwell et al., 2000). Stem rot results in wilted plants with greyish-green leaves (White, 1999), of which the internal pith of the lower stem becomes soft and disintegrated. Mycelial growth can be observed at lower nodes (White, 1999), and the stem exhibits a reddish discolouration internally when split open (White, 1999). Fusarium root rot symptoms appear as brownish lesions on the roots, and develop as early as the 6th leaf stage on the primary roots. When root rot continues to develop it can become severe on the adventitious roots (Sparks, 2016). This may reduce the length of primary roots and the quantity of secondary roots (Soonthornpoct et al., 2000).
Fusarium verticillioides is both an endophyte and a pathogen of maize (Yates et al., 1997).
The transition from endophytic stage to the pathogenicity stage is initiated by the single orthologous gene, SGE1, which regulates markedly different genes in different fungi (Brown
et al., 2014). In F. verticillioides, SGE1 regulates secondary metabolism and pathogenicity to
maize (Brown et al., 2014). Pathogenicity of F. verticillioides is also regulated by other genes such as the FvSO gene, which is required for vegetative growth and sporulation, fumonisin production and pathogenicity in F. verticillioides (Guo et al., 2015). Boudreau et al. (2013) showed that mutant F. verticillioides strains lacking a TPS1 gene, which encodes a putative trehalose-6-phosphate synthase, produced significantly less fumonisin and were less pathogenic than the non-mutant strain to maize. Moreover, Zhang et al. (2011) indicated that a mutant F. verticillioides strain lacking the FvMK1 gene, that regulates conidiation, pathogenesis, and fumonisin production, was non-pathogenic and did not colonise through wounding sites and failed to cause stem rot symptoms beyond the inoculation sites on maize stems. The F. verticillioides velvet gene FvVE1 also regulates stem rot symptom production and fumonisin synthesis in maize seedlings (Myung et al., 2012). Maize plants, grown from seeds inoculated with the FvVE1 deletion mutant, did not develop disease symptoms while plants grown from seeds inoculated with the F. verticillioides non-mutant strain developed disease symptoms (Myung et al., 2012).
Toxigenicity of F. verticillioides
Fumonisins in maize are produced by F. verticillioides, F. proliferatum (Nelson et al., 1983; Leslie and Summerell, 2006) and Aspergillus niger (Tiegh.) (Frisvad et al., 2011). Of these, F.
5
verticillioides is considered the most important producer of fumonisins worldwide (Desjardins,
2006; Picot et al., 2010). The mycotoxin was first isolated from maize collected in the Eastern Cape province of South Africa by Gelderblom et al. (1988). Since then, a family of fumonisin analogues have been characterised and grouped into fumonisins A, B, C and P series. Fumonisins produced by F. verticillioides consist primarily of the B series fumonisins, referred to as FB1, FB2 and FB3. Of these analogues, FB1 is the most important with a wide geographic distribution worldwide (Thiel et al., 1992; Musser and Plattner, 1997; Rheeder et al., 2002). Fumonisins can also be found in sorghum kernels (Shetty and Bhat, 1997), grapes (Vitis
vinifera L.) and raisins (Mogensen et al., 2010).
FB1 has been classified as a group 2B carcinogen by the International Agency for Research on Cancer (IARC), indicating that it is a possible carcinogen to humans (IARC, 1993). It inhibits the synthesis of sphingolipids (He et al., 2001; Riley et al., 2006), thereby interfering with the function of membrane proteins resulting in neural tube defects (NTD). Potential mechanisms for fumonisin hepatotoxicity and carcinogenicity include fatty acid accumulation and cell proliferation, oxidative stress, lipid peroxidation, peroxisome proliferation and disruption of the production of cellular lipids (Riley et al., 2006). Contamination of maize with FB1 has been correlated with the occurrence of human oesophageal cancer in the Eastern Cape province of South Africa (Rheeder et al., 1993) and the Cixian, Linxian and Shangqiu counties of China (Chu and Li, 1994; Yoshizawa et al., 1994), and a high incidence of NTD in infants whose mothers were consuming fumonisin-contaminated maize during pregnancy in Mexico (Missmer et al., 2006). However, it is known that maternal folic acid protects the foetus against NTD, and fortification of all enriched cereal products with folic acid reduces the occurrence of NTD (Green, 2002).
Exposure of animals to FB1 in feed results in a number of clinical symptoms. The most dramatic manifestation of maize feed contaminated with fumonisin is equine leukoencephalomalacia (ELEM), a neurotoxic disease of horses (Equus ferus caballus L.) and donkeys (Equus africanus asinus L.) that is induced by FB1 (Kellerman et al., 1990; Jovanović
et al., 2015). Contaminated feed containing FB1 levels as low as 8 µg/g FB1 exposes ponies
to an elevated risk of ELEM development (Wilson et al., 1992; Marasas, 1995). FB1 also induces porcine pulmonary oedema in pigs (Sus scrofa domestica L.) (Kriek et al., 1981; Harrison et al., 1990) and hepatocellular carcinoma, cholangiofibrosis and cholangiocarcinoma in rats (Rattus spp. Fischer de Waldheim) (Gelderblom et al., 1991). A number of species-specific effects have been experimentally induced by fumonisins on other animals such as immuno-suppression in chickens (Gullus domesticus L.), toxicity to broiler chicks and chicken embryos and nephrotoxicity in rabbits (Oryctolagus cuniculus L.) (Marasas, 1995).
6
Surveys conducted in Africa, North America, South America, Europe, the Middle East, South Asia, East Asia, Australia and New Zealand have shown that fumonisins occur in maize worldwide (Desjardins, 2006). The maximum tolerable daily intake for fumonisins recommended by the Joint Food and Agriculture Organization of the United Nations/World Health Organization (FAO/WHO) Expert Committee on Food Additives (JECFA) is 2 µg/g fumonisin per kg of human body weight per day in humans (WHO, 2002). The US Food and Drug Administration maximum levels for fumonisins in human food are set at 2 µg/g (FDA, 2001), while the European Union has maximum levels of 1 µg/g (EC, 2007) in food intended for direct human consumption. In South Africa, the maximum allowable limit for fumonisins in human food has recently been legislated at 2 µg/g (DOH, 2016). The maximum allowable limit in animal feed is, however, legislated in many countries (FAO, 2003).
Relationship between pathogenicity and toxigenicity
The occurrence of FER and the production of fumonisins are distinct aspects of F.
verticillioides infection in maize. Fumonisins have been shown to occur in maize ears without
visible FER symptoms (Munkvold et al., 1997; Sobek and Munkvold, 1999). Desjardins and Plattner (2000) indicated that fumonisin production is not required by F. verticillioides for maize ear infection and the development of FER. Glenn et al. (2008), however, demonstrated that F.
verticillioides from banana (Musa acuminata Colla); which neither produced fumonisins nor
caused disease to maize seedlings; became pathogenic and produced fumonisin on maize seedlings after the banana strain was transformed with the fumonisin biosynthetic gene (FUM) cluster. This pathogenesis was only demonstrated for foliar diseases on maize seedlings (Glenn et al., 2008).
Factors modulating Fusarium ear rot and fumonisin production in maize
Environmental conditions
Fusarium verticillioides and fumonisin production are predominant in maize grown in warmer
and drier climates, as well as cooler and moist climates (Ngoko et al., 2001; 2002; Santiago
et al., 2015). A model with non-linear, 3-dimensional Lorentzian equation indicated that
fumonisin production is largely influenced by climatic conditions such as temperature, rather than by rainfall, during the dough stage of kernel fill (Janse van Rensburg, 2012). Temperatures exceeding 35°C induce plant stress in maize and promote systemic infection of maize by F. verticillioides (Murillo-Williams and Munkvold, 2008). Decreasing soil moisture, and high soil and air temperatures during the later stages of ear maturity and dry down exacerbates silk cut, which causes lateral splits in the kernel pericarp (Murillo-Williams and Munkvold, 2008), thereby exposing the kernel tissues and embryo to F. verticillioides infection (Odvody et al., 1997). When grown outside their range of adaptation, hybrids appear to be
7
more susceptible to F. verticillioides infection and concomitant fumonisin production (Shelby
et al., 1994; Miller, 2001). However, hybrid x season and season x location interactions are
significant sources of variation for FER and fumonisin production, respectively (Venturini et
al., 2015).
Physio-chemical composition of kernels
Starch formation during kernel ripening progressively reduces water activity (aw) to levels that may trigger fumonisin production by F. verticillioides (Picot et al., 2010). The highest levels of FB1 production occur at the dent stage and the lowest at the blister stage (Fig. 1) (Warfield and Gilchrist, 1999). The pH and carbon:nitrogen ratio fluctuations during the course of maize ripening also modulate fumonisin production (Picot et al., 2010) and fumonisin biosynthesis in
F. verticillioides is repressed by high nitrogen levels and alkaline pH (Flaherty et al., 2003).
However, pH levels above 3.5 were found to enhance fungal growth in liquid cultures (Keller
et al., 1997). The optimum production of fumonisins occurs under relatively high oxygen
tensions and low pH due to organic acids produced by starch metabolism of F. verticillioides-colonised maize kernels (Miller, 2001).
The chemical composition of the pericarp in maize kernels plays a role in FER development and fumonisin production (Santiago et al., 2015). High ferulic acid content in the pericarp and aleurone tissues increases maize resistance to fungal infection (Bily et al., 2003). High levels of phenylpropanoids; such as trans- and cis-ferulic acid, p-coumaric acid and diferulate esters; have been shown to be associated with reduced FER and fumonisin production (Sampietro et al., 2013) due to their inhibitory effect on F. verticillioides (Santiago
et al., 2015). Studies by Venturini et al. (2015) indicated that flavonoids are an important
component in the resistance of maize to FER and fumonisin production. Significant differences between two isogenic hybrids (isohybrids), one with pigmentation in the pericarp and the other without pigmentation, were observed, with the pigmented kernels less affected by FER and fumonisin production than non-pigmented kernels.
Insects
Wounds produced by insects provide sites for infection of maize ears and stems by airborne or rain-splashed inoculum of F. verticillioides (Munkvold and Desjardins, 1997). Kernel damage caused by lepidopteran stem borers, such as the European corn borer, Ostrinia
nubilalis Hübner (Lepidoptera: Pyralidae), has been associated with high FER development
and fumonisin production in the USA (Munkvold et al., 1997). In South Africa, maize stem borers, mainly B. fusca, were also associated with a higher incidence of FER in maize (Flett and Van Rensburg, 1992).
8
Production practices
Production practices such as tillage, crop rotation, planting date, planting density and fertiliser treatment can influence FER and fumonisin production in maize. FER pathogens such as F.
verticillioides, F. proliferatum and F. subglutinans survive longer in fields with surface maize
residue when compared to fields with buried residues (Cotten and Munkvold, 1998). The primary inoculum in such residues can become airborne to infect maize ears and colonise kernels (Oren et al., 2003). Conservation tillage systems have been shown to sustain a higher diversity of Fusarium spp. than moldboard plough-based tillage systems (Steinkellner and Langer, 2004). No-till farming of maize following oats (Avena sativa L.) in Brazil resulted in higher fumonisin levels than maize conventional tillage following oats (Ono et al., 2011).
Late planting of maize, maize monoculture and crop rotation with other cereal crops in the same or in adjacent fields increase the build-up of F. verticillioides inoculum (Dowd, 2003). Late season planting affects kernel integrity, and thereby result in a significant increase in FER and FB1 production. In contrast, early planting consistently resulted in lower FER and FB1 levels in California and Hawaii compared to late planting dates (Parsons and Munkvold, 2012). High plant densities have increased maize kernel infection with F. verticillioides coupled with more FER compared to fields with lower plant densities (Blandino et al., 2008a). Plant densities that exceed agronomically recommended levels increase demand for water and nutrients, and such competition results in plant stresses that predispose plants to fungal infection and concomitant mycotoxin production (Bruns, 2003). Nitrogen deficiency caused increased fumonisin production in maize in Italy (Blandino et al., 2008b), and fumonisin levels have been shown to decrease with an increase in nitrogen fertilization rates (Ono et al., 2011).
Management of Fusarium ear rot and fumonisins
Biological methods
Several bacterial and fungal species suppress F. verticillioides in vitro and in planta. The bacterial maize endophyte Bacillus mojavensis releases Leu-7 surfactants suppress F.
verticillioides in vitro (Bacon and Hinton, 2002; Snook et al., 2009; Bacon and Hinton, 2011)
by forming micelles that result in high membrane-destabilising activity and solubilisation of the fungal membranes (Heerklotz and Seelig, 2001). As a seed dressing, however, its effectiveness was limited. It is believed that the production of fusaric acid by F. verticillioides prevents B. mojavensis from protecting seedlings against the fungus (Bacon et al., 2004; 2006). Bacillus amyloliquefaciens and Microbacterium oleovorans reduced F. verticillioides and fumonisin accumulation in maize kernels when applied as seed dressings at a concentration of 107 colony forming units ml−1 (Pereira et al., 2007). Seed dressing with
Bacillus cereus sensu lato has also reduced the incidence of FER and fumonisin production
9
Pseudomonas fluorescens effectively reduced the incidence of F. verticillioides and the level
of fumonisins in maize when used as a seed and spray treatment.
Acremonium zeae Gams & Sumner inhibited the growth of F. verticillioides in cultural tests
(Wicklow et al., 2005). This inhibition is due to the production of antibiotics, known as pyrrocidines A and B (Wicklow et al., 2005), which inhibit protein biosynthesis (Menninger, 1995). Trichoderma harzianum seed treatment reduced F. verticillioides infection by 58% and fumonisin production by 53% in maize in Italy (Ferrigo et al., 2014b), and can be used as an environmentally friendly method to reduce F. verticillioides infection (Sobowale et al., 2005; 2007). Root colonisation of maize by T. harzianum consistently reduced the pathogenicity of
F. verticillioides by activating the jasmonic acid (JA), salicylic acid (SA) and ethylene
(ET)-dependent defence mechanisms in maize plants (Ferrigo et al., 2014a).
Chemical methods
A number of fungicides were evaluated against F. verticillioides in the laboratory and in the field. Tebuconazole reduced the growth of F. verticillioides and F. proliferatum in vitro, but did not reduce the production of fumonisins (Marín et al., 2013). Maize seed treated with a protectant consisting of 35% Thriophate-methyl, 20% Thiram and 15% Diazinon, combined with an 80% Benomyl solution sprayed directly on the soil, tended to have similar F.
verticillioides stem infection as the F. verticillioides-uninoculated control following artificial
infection with F. verticillioides on the first internode (Schulthess et al., 2002). However, De Curtis et al. (2011) demonstrated that soil and foliar application of Tebuconazole and Tetraconazole, as well as the combination of Prochloraz + Cyproconazole with the insecticide Lambda-cyhalothrin, significantly reduced fumonisin production when applied to the soil and maize foliage. Lambda-cyhalothrin consistently reduced insect damage severity when applied alone, while fumonisin production was only reduced in 50% of cases (De Curtis et al., 2011). In Europe, more than 95% of the maize seeds planted are fungicide-treated. The treatments include amide, dithiocarbamate and pyrrole foliar fungicide sprays used against
Fusarium spp. for seed production (Meissle et al., 2010). Chemical elicitors, such as β-amino
butyric acid, benzothiadiazole, harpin protein, 2,6-dichloroisonicotinic acid and methyl jasmonate (MeJA), were unable to induce resistance to FER and fumonisin production (Small
et al., 2012b). The optimisation of elicitor application method, dosage rate and frequency, as
well as timing of application, could potentially increase plant response against F. verticillioides.
Cultural practices
Fusarium verticillioides infection and the subsequent accumulation of mycotoxins in maize
kernels can be controlled through cultural practices such as tillage, crop rotation and the application of fertilizers (Munkvold, 2003; Parsons and Munkvold, 2010). Removal of crop
10
residues from previous crops also reduced primary inoculum levels in the field (Dragich and Nelson, 2014). The survival of F. verticillioides can further be reduced by ploughing-in of surface residues (Cotten and Munkvold, 1998). Steinkellner and Langer (2004) indicated that moldboard ploughing resulted in a lower diversity of Fusarium spp. than the chisel plough and rotary tiller treatments, and the deeper the tillage the lower the number of surviving Fusarium spp. However, Flett et al. (1998) found that tillage practices applied prior to planting; such as rip-on-rowing followed by tiller, moldboard ploughing, shallow chisel, V-blade ploughing and disk-ploughing; had no effect on ear rots caused by Fusarium spp.
Crop rotation of maize with non-host crops of F. verticillioides breaks the life cycle of the pathogen and reduces FER and concomitant fumonisin production. Crop rotation with beans (Phaseolus spp. L.), groundnut (Arachis hypogaea L.) and potatoes (Solanum tuberosum L.) have all been demonstrated to reduce primary fungal inoculum (Atukwase et al., 2009; Dragich and Nelson, 2014). Sub-soiling in compacted soils minimises plant stress (Bruns, 2003) and reduces stress levels of maize in the field. Blandino et al. (2008b) also indicated that fumonisin contamination was highest in maize fields with nitrogen deficiencies.
Plant resistance
Progress made in genetic improvement of maize for F. verticillioides resistance has been limited. Breeding efforts to increase resistance to FER is uncommon due to the fact that ear rots rarely result in yield losses (Mesterházy et al., 2012). A hybrid named Mona was found to be resistant to F. verticillioides and F. proliferatum in Poland (Pascale et al., 2002), while Clements and White (2004), Afolabi et al. (2007) and Small et al. (2012a) found maize inbred lines in which FER was significantly reduced in the USA, Nigeria and South Africa, respectively in maize ears inoculated with F. verticillioides.
Post-harvest storage and treatment
The risk of fumonisin production increases in post-harvest maize kernels with high moisture levels (Warfield and Gilchrist, 1999). A moisture level exceeding 18% is favourable for F.
verticillioides growth and concomitant fumonisin production during storage (Kommedahl and
Windels, 1981). Subsistence farming practices; such as harvesting maize with a high moisture content and storing it in drums and tanks, as well as the non-control of weevils (Sitophilus
zeamais Motschulsky) and other insects (Ncube, 2008); may thus promote fungal growth and
mycotoxin production.
Standard storage procedures that prevent production of fumonisins in kernels, such as drying of maize kernels to moisture levels below 18% within 1-2 days after harvest, are recommended. Stored kernels should be aerated to reduce moisture content, and temperature set to 20°C (Joao and Lovato, 1999). Adjusting the combine harvester to avoid kernel damage
11
during harvesting further reduces fungal infection that could lead to mycotoxin production (Munkvold and Desjardins, 1997; Bruns, 2003). Bennett and Richard (1996) found that the gluten and fibre fractions of fumonisin-contaminated maize extracted by wet milling, rather than the starch, contained considerable amounts of fumonisins. These fractions, thus, require further decontamination before being used in animal feed. For subsistence farming systems, hand sorting mouldy from non-mouldy maize kernels prior to consumption is recommended to reduce fumonisin exposure (Desjardins et al., 2000; Van der Westhuizen et al., 2010).
Specific processes used in food preparation; such as nixtamalisation of maize dough (Palencia et al., 2003), the use of adsorbent clays (Baglieri et al., 2013), and heat treatment (Jackson et al., 1996); can reduce fumonisin levels in maize. The nixtamalisation of maize dough, however, may result in hydrolysed fumonisins that can be nearly as toxic as the unaltered FB1 (Murphy et al., 1996). Lu et al. (1997) found that non-enzymatic browning that occurs in the presence of a primary amine, a reducing sugar, and water at pH above 7, resulted in the removal of the primary amine group from the fumonisin molecule and concomitant reduction in detectable FB1 levels. Treatment of contaminated maize kernels with a combination of hydrogen peroxide and sodium bicarbonate reduced end-product toxicity (Park
et al., 1996). Reduction in the bioavailability of fumonisins can also be achieved through the
use of adsorbent clays, activated carbons and cholestyramine that have a high affinity for FB1. These compounds all tightly bind and immobilise FB1 in the gastro-intestinal tract of livestock and the immobilised FB1 is then excreted (Huwig et al., 2001; Solfrizzo et al., 2001). Cholestyramine has the best FB1 adsorption capacity, followed by activated carbon, bentonite and celite in rats (Solfrizzo et al., 2001). Fumonisins are fairly heat stable, but can be reduced by 16-100% when moist or dry maize kernels are heated at temperatures exceeding 160°C for 20-60 min (Jackson et al., 1997; Castelo et al., 1998; Katta et al., 1999).
THE AFRICAN MAIZE BORER BUSSEOLA FUSCA
Busseola fusca is the most important pest of maize in South Africa (Kfir and Bell, 1993; Kfir,
2000; 2002) while the spotted stem borer, Chilo partellus Swinhoe (Lepidoptera: Pyralidae), is of lesser economic importance (Kfir, 2002). The African pink stem borer, Sesamia calamistis Hampson (Lepidoptera: Noctuidae), is also considered a minor pest of maize in the country (Van den Berg, 1997; Van den Berg and Drinkwater, 2000). Some other Lepidoptera spp. that have attained pest status in South Africa include the black cutworm (Agrotis ipsilon Hufnagel), brown cutworm (Agrotis longidentifera Hampson), grey cutworm (Agrotis subalba Walker), and common cutworm (Agrotis segetum Denis & Schiffermüller), African bollworm (Helicoverpa
12
armigera Hübner), army worm (Spodoptera exempta Walker), and beet army worm
(Spodoptera exigua Hübner) (Annecke and Moran, 1982).
Life cycle of B. fusca
Pupal and moth stage
The life cycle of B. fusca consists of the adult female moth laying eggs that hatch into larvae and the larvae turning into pupae, from which adult moths will emerge to complete the cycle (Fig. 3). In South Africa B. fusca produces neither two nor three generations per season, but a combination of these (Mally, 1920; J.B.J. van Rensburg, personal communication). Three generations may occur if moths emerge early in spring (August). However, when a third generation of moths appear in the following spring, only two generations occur in that year. Pupae of the first summer generation occur from the end of December to January, and adults emerge to oviposit from the end of January to the first week in February on maturing plants (Mally, 1920). The larvae matures during March and April and some of these larvae pupate and emerge as moths, giving rise to a partial third generation. The larvae of these and the remainder of the second generation go into diapause in maize stubble (Mally, 1920). Adult moths mostly emerge at night and females release pheromones soon after emergence to attract males for mating (Harris and Nwanze, 1992). After mating female moths disperse in search of suitable host plants for oviposition over 3-4 successive nights (Harris and Nwanze, 1992).
Egg and larval stage
Busseola fusca moths oviposit their eggs on maize during the early vegetative stages.
Oviposition after tasselling occurs only when younger plants are not readily available (Van Rensburg et al., 1987b). The range of plant ages during which oviposition normally occurs might be extended in slow-growing hybrids and, therefore, slow-growing hybrids are prone to higher levels of infestation (Van Rensburg et al., 1989). Moths oviposit eggs in batches of 30-100 on the inner surfaces of leaf sheaths or on other smooth surfaces, such as the plant stem, during the early vegetative stages (Harris and Nwanze, 1992). The eggs are white initially, and darken with age, until they hatch into larvae in about 9 days (Bijlmakers, 1989; Harris and Nwanze, 1992; Robertson, 2000). The larvae then migrate to the whorl where feeding on the leaves takes place. Neonate larvae disperse within and between plants via ‘ballooning-off’, whereby first instar larvae cling onto silk threads enabling wind-facilitated movement (Kaufmann, 1983; Zalucki et al., 2002) during the pre-feeding movement phase (Zalucki et al., 2002; Calatayud et al., 2015). In addition to full ballooning, Calatayud et al. (2015) proposed that B. fusca neonate larvae migrate between plants by actively crawling, their orientation
13
toward host plants being guided by host plant-secreted volatiles. However, this form of larval migration has not yet been demonstrated under field conditions (Calatayud et al., 2015).
The majority of larvae pass through winter (diapause) in the base of the stem where they are sheltered from natural predators and adverse climatic conditions (Bijlmakers, 1989; Kfir, 1991; Harris and Nwanze, 1992; Robertson, 2000). Factors such as an increase in carbohydrate and decrease in protein and water are responsible for the induction of the diapause phase in B. fusca (Kfir, 1993a; 1993b). The larvae then emerge from diapause and pupate in the following spring between September and November (Kfir, 1991; Harris and Nwanze, 1992). Where maize is cultivated throughout the year, first generation moths that start their development process earlier tend to go through a third generation within the development year (Bijlmakers, 1989; Harris and Nwanze, 1992; Robertson, 2000). The maize crop planted during the summer in South Africa is usually exposed to a single generation of
B. fusca larvae (J.B.J. Van Rensburg, personal communication).
Busseola fusca host plants
Maize and sorghum are the best-known host crops for B. fusca (Kfir et al., 2002). Recent studies have indicated that the B. fusca host range has expanded to include sugarcane in small-scale farming systems in Ethiopia and southern Africa (Assefa et al., 2015). Sugarcane is grown adjacent to or in mixed systems with maize in small-scale farming systems, and this practice could have brought B. fusca in contact with sugarcane (Assefa et al., 2015). Moreover, this may have been facilitated by the presence of sugarcane as the only host during the winter when B. fusca is ordinarily in diapause in maize plants (Assefa et al., 2010; 2015). Other B. fusca host plants in South Africa include giant reed (Arundo donax L.), citronella grass (Cymbopogon nardus L.) Rendle and common wild sorghum (Sorghum arundinaceum Desv.) Stapf (Calatayud et al., 2014). Busseola fusca has also been found on Guinea grass (Panicum maximum Jacq.), Napier grass (Pennisetum purpureum Shumach.), lemon grass (Cymbopogon giganteus Chiov.), pearl millet, and bigleaf bristle grass (Setaria megaphylla Steud.) Duran & Schinz in East Africa (Calatayud et al., 2014). Busseola phaia Bowden, B.
segeta Bowden and B. nairobica Le Rü, which closely resemble B. fusca, occur on wild host
plants around cereal crops. However, B. fusca accounts for only about 14% of all the Busseola spp. collected on wild hosts (Calatayud, et al., 2014).
Behaviour of B. fusca moths and larvae on host plants
Oligophagous insects such as B. fusca have a narrow host range due to strong selectivity (Bernays and Chapman, 1994). The gravid female moth is primarily responsible for host plant recognition and selection (Van den Berg, 2006). Recognition and colonisation is based on the interaction between the insect’s sensory systems and the physicochemical characteristics of
14
its immediate environment (Udayagiri and Mason, 1995; Calatayud et al., 2006). The sensory equipment of the moth consists of multi-porous chemoreceptor and mechanoreceptor sensillae on the antennae which receive plant volatiles, and uniporous gustatory sensilla on the antennae, tarsi and ovipositor that probes the plant surface (Calatayud et al., 2006).
Lepidopteran insects use various sensory cues (stimuli) to locate and accept host plants (Bernays and Chapman, 1994). Visual and olfactory cues play an essential role before the moth lands on the plant (Rojas and Wyatt, 1999), whereas olfactory and tactile cues are thought to be more crucial after landing (Hora and Roessingh, 1999). The moth acquires information on the quality and suitability of the plant for colonisation upon landing (Renwick and Chew, 1994). The final behavioural sequence leading to acceptance or rejection of the site for oviposition depends mainly on contact cues consisting of both physical factors such as pubescence and surface texture, and chemical cues such as volatile and surface chemicals present (Hora and Roessingh, 1999).
The host plant selection by insects is usually divided into ‘host plant finding’ and ‘host plant acceptance’ (Finch and Collier, 2000). During host plant finding, the searching moths avoid landing on brown surfaces such as soil, but lands indiscriminately on green objects such as leaves of host plants. Landing on green plant tissue is referred to as ‘appropriate landings’, whereas landing on non-host plants is called ‘inappropriate landings’ (Finch and Collier, 2000). After landing, the female typically sweeps her ovipositor on the plant surface, simultaneously touching it with the tips of her antennae, and then oviposit (Renwick and Chew, 1994; Calatayud et al., 2008). This behaviour is more frequently observed in maize and sorghum, indicating that both antennal and ovipositor receptors are used by the female moths to evaluate the plant surface before ovipositing (Calatayud et al., 2008). Females recognise their preferred hosts only after landing where tactile and contact-chemoreception cues from the plants play a major role in oviposition decisions (Calatayud et al., 2008). Busseola fusca moths have a preference for thick-stemmed plants for oviposition (Van Rensburg and Van den Berg, 1990), and high plant density promotes oviposition (Van Rensburg and Van den Berg, 1990).
The behavioural steps that lead to oviposition by a female moth generally follow a sequential pattern consisting of searching, orientation, encounter, landing, surface evaluation, and acceptance (Renwick and Chew, 1994). Visual or chemical cues, and most probably a combination of both, may trigger the general orientation of a B. fusca moth toward the host plant (Renwick and Chew, 1994; Calatayud et al., 2008). The insect can visually locate an upright plant stem by flying directly towards it and then landing. Shape is another visual cue that may play an important role in the general orientation towards the plant in nocturnally active lepidopterans, such as B. fusca (Renwick and Chew, 1994). The physiological status of the insect, such as its age and duration of deprivation of a suitable ovipositing site, also plays a role in host plant selection (Barton-Browne, 1993).
15
Stem borer larvae produce oral secretions such as regurgitated feed and those of salivary glands during feeding. These secretions contain herbivore-associated molecular patterns (HAMPs), which affect plant responses to herbivores as well pathogenic fungi or bacteria through the SA/JA cross-talk (Alborn, 1997; Musser et al., 2002; Mithöfer and Boland, 2008; Schäfer et al., 2011; Louis et al., 2013). Herbivore-associated contact due to feeding, oviposition and crawling also induce plant defences (Mithöfer et al., 2005; Hilker and Meiners, 2006; Peiffer et al., 2009; Bricchi et al., 2010). Another important stem borer behavior, defecation, also plays a role in plant-herbivore-pathogen interactions (Ray et al., 2015). Lepidopteran pests, such as the fall armyworm Spodoptera frugiperda Smith (Lepidoptera: Noctuidae), deposit substantial amounts of frass (excreta) within the enclosed whorl tissue surrounding their feeding site, where it remains for long periods of time (Ray et al., 2015). The frass is composed of molecules derived from the host plant, the insect itself, and associated microbes; and it provides abundant cues that alter plant defence responses (Ray et al., 2015). It has been observed that proteins from S. frugiperda frass induced wound-responsive defence genes in maize (Ray et al., 2015). Elicitation of pathogen defences by frass proteins was found to increase herbivore damage and reduce southern maize leaf blight caused by Cochliobolus
heterostrophus (Drechsler) (Shanmugam et al., 2010; Ray et al., 2015). Moreover, S.
frugiperda frass reduced the accumulation of maize herbivore defence gene transcripts and
JA levels, while elevating the abundance of pathogen defence gene transcript (Ray et al., 2015).
Damage caused to maize plants by stem borers
Busseola fusca larvae cause damage to all maize plant parts (Calatayud et al., 2014). First
generation B. fusca larvae feed within the whorls until the third instar thus destroying the growing point of the plant (dead-heart) (Van Rensburg et al., 1988a). They thereafter enter the plant stem (Van Rensburg et al., 1987b) where tunnelling results in extensive damage to internal stem tissue (Van Rensburg et al., 1988a). Busseola fusca larvae also cause direct damage to maize ears, and damage to plants after tasselling has the most important influence on yield (Van Rensburg et al., 1988a). The number of larvae per plant is, however, a poor indicator of expected yield loss (Van Rensburg et al., 1988d). After ballooning-off and feeding on whorl tissue, third instars bore into the stem and older instars may also migrate in search of more suitable host plants (Kaufmann, 1983). Only one larva usually feed per stem due to the cannibalistic nature of B. fusca larvae (Robertson, 2000). There are usually six larval instar stages, and larvae mature in about 35 days and pupate inside the stem for about 2-3 weeks. They make small exit holes before pupating to enable the emergence of the adult moth (Bijlmakers, 1989; Harris and Nwanze, 1992; Robertson, 2000).
16
Factors modulating survival and growth of B. fusca larvae
Environmental conditions
Weather conditions such as high wind speed affect moth flight activity and ballooning-off movement of larvae, while heavy rainfall results in drowning of neonate larvae (Van Rensburg
et al., 1988b; Zalucki et al., 2002). Low rainfall, in contrast, results in reduced seasonal
abundance of B. fusca moths (Van Rensburg et al., 1987a; Ebenebe et al., 2000a), while the daily flight activity and survival of moths is enhanced by cool and humid conditions (Van Rensburg et al., 1987a). High levels of infestations can occur during years with adequate rainfall after older larval instars have penetrated the stem where they are protected from the adverse effects of rainfall (Van Rensburg et al., 1987a).
Production practices
Increased maize plant population enhances both the rates of dispersal and the survival of B.
fusca larvae (Van Rensburg et al., 1988c; Van den Berg et al., 1991). A reduction in row width
puts the adjacent plant rows within reach of more larvae through migration, resulting in increased plant damage (Van Rensburg et al., 1988c). The infestation potential of stem borers in maize is increased by the incidence of tillering and the availability of leaf sheaths that are suitable for oviposition (Van Rensburg and Van den Berg, 1990).
Management of maize stem borers
Biological control
Parasitoids that attack B. fusca larvae can be used as biological control agents. Cotesia
flavipes Cameron (Hymenoptera: Braconidae) is the only exotic parasitoid that has
established itself on maize in mainland Africa (Kfir et al., 2002). The short seasonal duration of maize farming provides only a 3-month habitat for stem borers and their natural predators. This lack of habitat stability results in the failure of parasitoids to establish and control stem borers (Hall and Ehler, 1979). Parasitoids could, in theory, have a better chance of establishing themselves in subsistence farming systems where maize is grown all year round, under irrigation or in small gardens. A parasitoid that has a wide host range of target species also has a better chance of establishment in a new area compared to a parasitoid with a narrow host range (Kfir et al., 2002). Climatic compatibility further influences parasitoid establishment. The temperate climate of South Africa has not been conducive to the establishment of parasitoids from tropical and sub-tropical regions (Skoroszewski and Van Hamburg, 1987; Kfir, 1994).
A positive relationship between any component of fitness of a species, and either numbers or density of conspecifics (Allee effect) (Fig. 4) (Berec et al., 2006), might explain why natural predators fail to establish as biological control agents (Hopper and Roush, 1993). Release of
